[PVE-User] Node update that went awry...

David Thompson david at digitaltransitions.ca
Fri Feb 1 02:43:18 CET 2013


Hi all,

I have a 2 node system running (non HA) on an intel modular server. The configuration is as follows:

3 X 140GB Hard Drives split in RAID 5 for proxmox boot systems
4 X 900GB Hard Drives (with lun share key) for proxmox KVM images.

I followed the extending local container storage article here:
http://pve.proxmox.com/wiki/Extending_Local_Container_Storage
and this one here:
http://pve.proxmox.com/wiki/Intel_Modular_Server#Pool2:_Shared_LVM_storage_for_KVM_guests

And everything has been going swimmingly. I decided today to update it from the .16 kernel to the .17 kernel. I know if it ain't broke don't fix it. One day I will follow that golden rule.
The update on the first node went well. I migrated all the VM's and containers from the first node to the second node before updating and everything is running fine. 
When I went to restart node1, it wouldn't boot off the new kernel.
------------------------------------------------------------------------------
The error that I get on the boot screen is:
	Booting 'Proxmox Virtual Environment GNU/Linuc, with'
Loading Linux 2.6.23-16.pve
error: file not found
Loading Initial ramdisk ...
error: you need to boot both default and fallback entries
------------------------------------------------------------------------------

I was then able to choose the 2.6.32-11-pve kernel to boot off of and all appeared to be fine, however, when ever I try and update it (aptitude and or apt-get) shows nothing to update. So I manually updated the kernels with the following:
	apt-get install pve-kernel-2.6.32-16-pve
	apt-get install pve-headers-2.6.32-16-pve

It installed, but I got the following error 

Generating grub.cfg ...
/usr/sbin/grub-probe: error: Couldn't find PV pv1. Check your device.map.

-->> I have read from googling that removing /boot/grub/grub.cfg will regenerate the device.map file with an apt-get -f install, but I don't want to do any damage to this. I am also currently afraid to reboot this node server in case it doesn't come back either with the new proper current kernel and or the older kernel.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
root at node1:/boot# apt-get install pve-kernel-2.6.32-16-pve
Reading package lists... Done
Building dependency tree       
Reading state information... Done
The following NEW packages will be installed:
  pve-kernel-2.6.32-16-pve
0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
Need to get 0 B/31.5 MB of archives.
After this operation, 0 B of additional disk space will be used.
Selecting previously deselected package pve-kernel-2.6.32-16-pve.
(Reading database ... 32700 files and directories currently installed.)
Unpacking pve-kernel-2.6.32-16-pve (from .../pve-kernel-2.6.32-16-pve_2.6.32-82_amd64.deb) ...
Setting up pve-kernel-2.6.32-16-pve (2.6.32-82) ...
update-initramfs: Generating /boot/initrd.img-2.6.32-16-pve
Generating grub.cfg ...
/usr/sbin/grub-probe: error: Couldn't find PV pv1. Check your device.map.

root at node1:/boot# apt-get install pve-headers-2.6.32-16-pve
Reading package lists... Done
Building dependency tree       
Reading state information... Done
The following NEW packages will be installed:
  pve-headers-2.6.32-16-pve
0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
Need to get 6,310 kB of archives.
After this operation, 0 B of additional disk space will be used.
Get:1 http://download.proxmox.com/debian/ squeeze/pve pve-headers-2.6.32-16-pve amd64 2.6.32-82 [6,310 kB]
Fetched 6,310 kB in 12s (492 kB/s)                                                                                                                                                                                   
Selecting previously deselected package pve-headers-2.6.32-16-pve.
(Reading database ... 35125 files and directories currently installed.)
Unpacking pve-headers-2.6.32-16-pve (from .../pve-headers-2.6.32-16-pve_2.6.32-82_amd64.deb) ...
Setting up pve-headers-2.6.32-16-pve (2.6.32-82) ...
root at node1:/boot# uname -a
Linux node1 2.6.32-11-pve #1 SMP Wed Apr 11 07:17:05 CEST 2012 x86_64 GNU/Linux
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

The other issue is when I got it back up and running, I tried to migrate one of the windows 2008 servers from the KVMStorage on node2 back to node1 and I got the following error:

Jan 31 14:52:43 starting migration of VM 10000 to node 'node1' (192.168.110.2)
Jan 31 14:52:43 copying disk images
Jan 31 14:52:43 starting VM 10000 on remote node 'node1'
Jan 31 14:52:45 starting migration tunnel
Jan 31 14:52:45 starting online/live migration on port 60000
Jan 31 14:52:47 migration status: active (transferred 68827161, remaining 4234821632), total 4312137728)
------- [ SNIP SNIP SNIP] -----
Jan 31 14:54:22 migration status: active (transferred 3236421728, remaining 40189952), total 4312137728)
Jan 31 14:54:22 migration speed: 42.23 MB/s
Jan 31 14:54:22 migration status: completed
Jan 31 14:54:23 ERROR: command '/usr/bin/ssh -o 'BatchMode=yes' root at 192.168.110.2 qm resume 10000 --skiplock' failed: exit code 2
Jan 31 14:54:25 ERROR: migration finished with problems (duration 00:01:43)
TASK ERROR: migration problems



Sorry, this is a long email with a couple of issues that have occurred from what was supposed to be a simple update. 
Thanks to anyone that can help out with my problem...

__________
David




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://pve.proxmox.com/pipermail/pve-user/attachments/20130131/f4d7ab2e/attachment-0013.html>


More information about the pve-user mailing list